Matches in SemOpenAlex for { <https://semopenalex.org/work/W2890123813> ?p ?o ?g. }
Showing items 1 to 75 of
75
with 100 items per page.
- W2890123813 endingPage "2004" @default.
- W2890123813 startingPage "1994" @default.
- W2890123813 abstract "We present SplineNets, a practical and novel approach for using conditioning in convolutional neural networks (CNNs). SplineNets are continuous generalizations of neural decision graphs, and they can dramatically reduce runtime complexity and computation costs of CNNs, while maintaining or even increasing accuracy. Functions of SplineNets are both dynamic (i.e., conditioned on the input) and hierarchical (i.e.,conditioned on the computational path). SplineNets employ a unified loss function with a desired level of smoothness over both the network and decision parameters, while allowing for sparse activation of a subset of nodes for individual samples. In particular, we embed infinitely many function weights (e.g. filters) on smooth, low dimensional manifolds parameterized by compact B-splines, which are indexed by a position parameter. Instead of sampling from a categorical distribution to pick a branch, samples choose a continuous position to pick a function weight. We further show that by maximizing the mutual information between spline positions and class labels, the network can be optimally utilized and specialized for classification tasks. Experiments show that our approach can significantly increase the accuracy of ResNets with negligible cost in speed, matching the precision of a 110 level ResNet with a 32 level SplineNet." @default.
- W2890123813 created "2018-09-27" @default.
- W2890123813 creator A5014927177 @default.
- W2890123813 creator A5030249562 @default.
- W2890123813 date "2018-01-01" @default.
- W2890123813 modified "2023-09-29" @default.
- W2890123813 title "SplineNets: Continuous Neural Decision Graphs" @default.
- W2890123813 hasPublicationYear "2018" @default.
- W2890123813 type Work @default.
- W2890123813 sameAs 2890123813 @default.
- W2890123813 citedByCount "7" @default.
- W2890123813 countsByYear W28901238132019 @default.
- W2890123813 countsByYear W28901238132020 @default.
- W2890123813 countsByYear W28901238132021 @default.
- W2890123813 crossrefType "proceedings-article" @default.
- W2890123813 hasAuthorship W2890123813A5014927177 @default.
- W2890123813 hasAuthorship W2890123813A5030249562 @default.
- W2890123813 hasConcept C102634674 @default.
- W2890123813 hasConcept C11413529 @default.
- W2890123813 hasConcept C119857082 @default.
- W2890123813 hasConcept C126255220 @default.
- W2890123813 hasConcept C134306372 @default.
- W2890123813 hasConcept C154945302 @default.
- W2890123813 hasConcept C165464430 @default.
- W2890123813 hasConcept C199360897 @default.
- W2890123813 hasConcept C2777735758 @default.
- W2890123813 hasConcept C33923547 @default.
- W2890123813 hasConcept C41008148 @default.
- W2890123813 hasConcept C45374587 @default.
- W2890123813 hasConcept C50644808 @default.
- W2890123813 hasConcept C5274069 @default.
- W2890123813 hasConceptScore W2890123813C102634674 @default.
- W2890123813 hasConceptScore W2890123813C11413529 @default.
- W2890123813 hasConceptScore W2890123813C119857082 @default.
- W2890123813 hasConceptScore W2890123813C126255220 @default.
- W2890123813 hasConceptScore W2890123813C134306372 @default.
- W2890123813 hasConceptScore W2890123813C154945302 @default.
- W2890123813 hasConceptScore W2890123813C165464430 @default.
- W2890123813 hasConceptScore W2890123813C199360897 @default.
- W2890123813 hasConceptScore W2890123813C2777735758 @default.
- W2890123813 hasConceptScore W2890123813C33923547 @default.
- W2890123813 hasConceptScore W2890123813C41008148 @default.
- W2890123813 hasConceptScore W2890123813C45374587 @default.
- W2890123813 hasConceptScore W2890123813C50644808 @default.
- W2890123813 hasConceptScore W2890123813C5274069 @default.
- W2890123813 hasLocation W28901238131 @default.
- W2890123813 hasOpenAccess W2890123813 @default.
- W2890123813 hasPrimaryLocation W28901238131 @default.
- W2890123813 hasRelatedWork W1686810756 @default.
- W2890123813 hasRelatedWork W1861492603 @default.
- W2890123813 hasRelatedWork W1902237438 @default.
- W2890123813 hasRelatedWork W2097117768 @default.
- W2890123813 hasRelatedWork W2108598243 @default.
- W2890123813 hasRelatedWork W2117539524 @default.
- W2890123813 hasRelatedWork W2194775991 @default.
- W2890123813 hasRelatedWork W2612445135 @default.
- W2890123813 hasRelatedWork W2947630742 @default.
- W2890123813 hasRelatedWork W2949462966 @default.
- W2890123813 hasRelatedWork W2951532585 @default.
- W2890123813 hasRelatedWork W2953055086 @default.
- W2890123813 hasRelatedWork W2962935523 @default.
- W2890123813 hasRelatedWork W2962944050 @default.
- W2890123813 hasRelatedWork W2963163009 @default.
- W2890123813 hasRelatedWork W2963393494 @default.
- W2890123813 hasRelatedWork W2963403868 @default.
- W2890123813 hasRelatedWork W2963420686 @default.
- W2890123813 hasRelatedWork W2964308564 @default.
- W2890123813 hasRelatedWork W2970107519 @default.
- W2890123813 hasVolume "31" @default.
- W2890123813 isParatext "false" @default.
- W2890123813 isRetracted "false" @default.
- W2890123813 magId "2890123813" @default.
- W2890123813 workType "article" @default.